Long-Term Treatment of Chronic Obstructive Pulmonary Disease with Salmeterol and the Additive Effect of Ipratropium

Long-term treatment of chronic obstructive pulmonary disease with salmeterol and the additive effect of ipratropium

J.A. van Noord*, D.R.A.J. de Munck**, Th.A. Bantje***, W.C.J. Hop+, M.L.M. Akveld++, A.M. Bommer++

Long-term treatment of chronic obstructive pulmonary disease with salmeterol and the Depts of Respiratory Diseases, *Atrium additive effect of ipratropium. J.A. van Noord, D.R.A.J. de Munck, Th.A. Bantje, W.C.J. medisch centrum, Heerlen, **St Joseph Hospital, Veldhoven and ***Ignatius Hos- Hop, M.L.M. Akveld, A.M. Bommer. #ERS Journals Ltd 2000. + ABSTRACT: The efficacy and safety of salmeterol alone was compared with the pital, Breda, Dept of Epidemiology and Biostatistics, Erasmus University, Rotter- combination of salmeterol plus ipratropium and with placebo during long-term dam and ++GlaxoWellcome, Zeist, the treatment in patients with stable chronic obstructive pulmonary disease. In addition, Netherlands. the single-dose effect in response to the first dose of treatment was studied over 12 h. The patients (n=144; age 647 yrs, forced expiratory volume in one second (FEV1) Correspondence: J.A. van Noord, Dept of 4411% pred) participated in a three-centre double-blind double-placebo parallel Respiratory Diseases, Atrium medisch group study and were randomized after a run-in period of 2 weeks to receive either centrum, Henri Dunantstraat 5, 6419 PC salmeterol 50 mg b.i.d., salmeterol 5 mg b.i.d. plus ipratropium 40 mg q.i.d. or placebo Heerlen, The Netherlands. Fax: 31 for a period of 12 weeks. 455767534. The single-dose study demonstrated that salmeterol produced a significant increase Keywords: Chronic obstructive in FEV1 (peak of 7% pred) and specific airway conductance (sGaw) (maximum of 60% pulmonary disease baseline) for $12 h. The combination of salmeterol plus ipratropium elicited a greater ipratropium bronchodilator response (11% and 94% increases respectively) than salmeterol alone salmeterol during the first 6 h after inhalation. During treatment there were significant improvements in daytime symptom scores and morning peak expiratory flow in both Received: March 19 1998 the salmeterol and the salmeterol plus ipratropium groups (p<0.001), with an Accepted after revision November 7 1998 associated decrease in the use of rescue salbutamol. Improvements in FEV1 and sGaw were greater in the salmeterol plus ipratropium group than in the patients receiving only salmeterol. Thirty-five patients had an exacerbation; 11 (23%) in the salmeterol group (versus placebo NS), six (13%) in the salmeterol plus ipratropium group (versus placebo p<0.01) and 18 (36%) in the placebo group. In conclusion, in patients with severe stable chronic obstructive pulmonary disease, long-term treatment with either salmeterol alone or salmeterol plus ipratropium is safe and effective. There was added benefit from the combination therapy in terms of improvement in airways obstruction, but not for improvement in symptom control or need for rescue salbutamol. Eur Respir J 2000; 15: 878±885.

Chronic obstructive pulmonary disease (COPD) is defi- salmeterol produced an improvement in lung function and ned as a disease state characterized by the presence of daily symptom scores and was associated with a clinically airflow obstruction, due to chronic bronchitis or emphy- significant gain in quality of life [7±10]. Recently MAHLER sema, and is generally progressive but may be partially et al. [11], comparing inhaled salmeterol and ipratropium reversible [1]. The goals of pharmacotherapy in this dis- in patients with COPD over a 12-week period, found that order are to induce bronchodilation and facilitate expec- salmeterol was better than ipratropium at improving lung toration in order to improve symptoms, prevent recurrent function. To date, long-term studies on combination ther- exacerbations and so enhance the quality of life [2]. In apy with salmeterol and ipratropium bromide are not patients with stable COPD, the spirometric response to available. A number of trials have shown that, in patients bronchodilators such as b2-agonists, anticholinergic drugs with stable COPD, the combination of short-acting b2- and methylxanthines is at best very modest. However, even agonists and ipratropium may be beneficial [12], and the in the absence of significant bronchodilation, an improve- combination of salmeterol and ipratropium may have ment in symptoms and exercise tolerance can be demon- additive effects and long-term benefits. strated [3]. This study was designed to compare the efficacy and Salmeterol xinofoate, a long-acting b2-agonist, has been safety of salmeterol either alone or in combination with shown in single-dose studies to induce a spirometric im- ipratropium bromide with that of placebo during a 12-week provement over a 12-h period in patients with COPD [4± treatment period in patients with stable COPD. In addition, 6]. In addition, three studies in patients with COPD over the single-dose effect in response to the first dose of the periods of 1±3 months have found that treatment with treatment was studied over a period of 12 h. SALMETEROL VERSUS SALMETEROL PLUS IPRATROPIUM IN COPD 879

Methods Measurements First dose of treatment. Airway resistance (R ), specific Patients aw airway conductance (sGaw), FEV1 and FVC were obtained Patients had to meet the following criteria to be recruited at baseline and 0.5, 1, 2, 3, 4, 5, 6, 8, 10 and 12 h after in- for the study: 1) current or exsmokers with a smoking his- halation of the trial drugs. FEV1 and FVC were recorded at tory equivalent to 10 pack-yrs and with COPD according to the mouth using a pneumotachograph with electronic inte- American Thoracic Society criteria [1]; 2) aged 40±75 yrs; gration. The best of three adequate measurements was 3) no change in medication for COPD in the preceding 6 retained. Raw and sGaw were measured in a pressure-comp- weeks and no major changes in smoking habits during the ensated integrated flow plethysmograph (SensorMedics last 6 months; and 4) a forced expiratory volume in one 2800 Autobox; SensorMedics Corp., Yorba Linda, CA, second (FEV1)of#75% of the predicted value after in- USA) as the slopes of the chords between inspiratory and halation of 200 mg salbutamol via metered dose inhaler expiratory flows of 0.5 L.s-1 at a respiratory frequency of (MDI). Exclusion criteria included the following: a his- 0.5 Hz. The means of three measurements are reported. In tory of asthma, allergic rhinitis, atopy or a total blood addition blood pressure and cardiac frequency were obtain- eosinophil count of >500 cells.mm-3; respiratory disease ed at baseline and 1, 2, 3, 4, 5, 6, 8 and 12 h after inhalation. other than COPD; any clinically significant concurrent Twelve-week treatment. Diary card. During the run-in and disease; and oxygen therapy. treatment periods, all patients completed a daily diary card, recording their morning and evening peak expiratory flow (PEF), daytime and night-time symptoms and use of rescue Study design salbutamol. PEF was measured using a Personal Best peak flow meter (Healthscan Products Inc., Cedar Grove, NJ, The study had a randomized, double-blind double-pla- USA) between 07:00 and 08:00 h before taking the study cebo, parallel group design and was performed at three medication and between 19:00 and 20:00 h. The best of centres in the Netherlands. The study was approved by the three measurements was retained. The use of rescue medi- Medical Ethics Committees of all the participating hospi- cation within 4 h of the measurements was also recorded. tals. Written informed consent was obtained from all pa- The following symptom scores were used. Daytime symp- tients before any study procedure was undertaken. The tom score: 0: no symptoms at rest or on exertion; 1: no study had a run-in period of 2 weeks, a treatment period of symptoms at rest but symptoms on moderate exertion, e.g. 12 weeks and a follow-up period of 2 weeks. There were walking quickly, climbing stairs, rushing out to work; 2: no six scheduled visits to the clinic: at the start of the run-in symptoms at rest but symptoms on mild exertion, e.g. get- period (visit 1), at the start of treatment (visit 2), after 4, 8 ting dressed or washed; 3: minimal symptoms at rest, e.g. while sitting reading or watching the television; 4: moder- and 12 weeks of treatment (visits 3±5) and 2 weeks after ate symptoms at rest, e.g. while sitting reading or watching stopping treatment (visit 6) for follow-up. At the start of the the television; and 5: severe symptoms at rest, i.e. patient run-in period, all b -agonists and anticholinergics were 2 unable to carry out any activity requiring exertion. This withdrawn. The patients continued to take the permitted daytime symptom score has been used previously in COPD medication for their COPD in stable doses, including meth- [7]. Night-time symptom score: 0: no symptoms during ylxanthines, oral steroids and inhaled steroids up to 2,000 . -1 . -1 the night; 1: symptoms causing patient to wake once or mg day and fluticasone propionate up to 1,000 mg day . wake early; 2: symptoms causing patient to wake twice or Patients were given salbutamol as rescue medication. At the more (including waking early); 3: symptoms causing end of the run-in period (visit 2), patients were eligible for patient to be awake most of the night; and 4: symptoms so randomization if they fulfilled the following inclusion cri- severe that patient did not sleep at all. This five-point teria: 1) an FEV1 of #65% pred and $0.75 L at visit 1 or 2, symptom score for night-time use has also been used bronchodilators having been withheld for an appropriate previously [9, 13]. period, i.e. $4 h for short-acting bronchodilators, $24 h for Clinic visits. At each scheduled visit, details of adverse long-acting b2-agonists and 36 h for methylxanthines; 2) an events, exacerbations and withdrawals were recorded and FEV1/forced vital capacity (FVC) ratio of #60% at visit 1 or Raw,sGaw,FEV1 and FVC were measured. All lung func- 2; and 3) A daytime symptom score of $2onatleast4outof tion measurements were performed between 08:00 and 7 days during the run-in period (see Diary card section). 12:00 h after the morning dose of the study medication. At The eligible patients were randomly assigned to treat- each visit, the response to salbutamol 200 mgwasmeas- ment with 1) salmeterol 50 mg plus ipratropium bromide ured 15 min after inhalation. An exacerbation was defined matched placebo; 2) salmeterol 50 mg plus ipratropium as an increase in symptoms requiring either a course of oral bromide 40 mg; or 3) salmeterol-matched placebo plus ip- corticosteroids or antibiotics or a hospital admission. This ratropium bromide-matched placebo. After randomization, change in medication was at the investigator'sdiscretion. the first dose of the study medication was given and lung function was measured for 12 h. Subsequently, the patient Statistical analysis was treated with the same regimen for a period of 12 weeks. Salmeterol or placebo matched to salmeterol was Comparison of percentages between groups was done inhaled b.i.d. at 08:00 and 22:00. Ipratropium or placebo using Fisher's exact test. For the single dose part of the matched to ipratropium was inhaled q.i.d. at the following study, the change (per cent) from baseline of FEV1 and time points 08:00, 13:.00, 17:00, and 22:00 h. All medi- sGaw was determined at each timepoint. To compare cations were inhaled from an MDI attached to an aerosol treatment groups, the area under the curve (AUC) for these chamber (Volumatic1;GlaxoWellcome,UK). changes was calculated using the trapezoid rule. In a few 880 J.A. VAN NOORD ET AL. patients (n=9), all but one after placebo, the single-dose Table 1. ± Baseline characteristics of the randomized study had to be interrupted due to shortness of breath. In patients these patients the last observation made was carried forward Salmeterol Salmeterol plus Placebo to determine the AUC. Analysis of variance (ANOVA) was ipratropium used to compare mean AUCs between treatment groups. The same test was used to compare treatment groups at the Patients n 47 47 50 different time points. Sex M/F % 89/11 88/12 86/14 The change from baseline (per cent) of the lung function Current smoker % 49 57 58 test outcomes at the visits after 4, 8 and 12 weeks of treat- Age yrs 656 637 637 ment were compared between and within groups using Visit 1 (screening) FEV1 L 1.30.4 1.40.4 1.30.4 repeated measures ANOVA using the BioMedical Data Reversibility % Processing (BMDP) Statistical Software package [14]. To % baseline 1310 1611 159 evaluate diary card outcomes (symptom scores, PEF, and % FEV1 pred 64 75 63 use of rescue medication), the days during treatment were Visit 2 (baseline) grouped according to the time at which visits at the clinic FEV1 L 1.20.4 1.20.4 1.10.3 occurred, i.e. weeks 1±4, 5±8 and 9±12. Mean values du- FEV1 % pred 4210 4112 3810 ring these treatment periods were determined, and com- FVC L 2.80.8 3.00.8 2.80.7 FEV1/FVC % 438 429 419 pared to baseline values obtained during the run-in period. . -1. For individual patients, values obtained in a particular Raw kPa L s 0.630.23 0.650.23 0.720.30 sGaw kPa.L-1.s 0.350.18 0.330.13 0.290.09 treatment period were considered evaluable only if the Concurrent number of diary card days completed by the patient was medication $7. Repeated-measures ANOVA was used to compare b2-agonists* n 45 52 51 between- and within- group changes from baseline. The Anticholinergics* n 23 17 22 limit for statistical significance was set at two-sided p= Methylxanthines n 9 5 4 0.017 (i.e. 0.05/3) in the pairwise comparison of treatment Inhaled groups, according to Bonferroni's principle [15]. In the corticosteroids n 38 35 38 case of withdrawal, all data obtained prior to that point were Oral included in all analyses. corticosteroids n 5 5 1 Mucolytics n 16 12 18 Results Data are meanSD. *: withdrawn at visit 1. F: female; M: male. Of the 177 patients recruited for the study, 29 were FEV1: forced expiratory volume in one second; %FEV1 pred: excluded before randomization because of ineligibility, and percentage of predicted FEV1; FVC: forced vital capacity; % four were excluded from analysis due to protocol viola- pred: percentage of predicted value; Raw: airway resistance; tions. Of the remaining 144 patients, 47 received salme- sGaw: specific airway conductance. terol, 47 salmeterol plus ipratropium and 50 placebo. The characteristics of these patients are given in table 1. The terol, there was a maximum increase in sGaw of 607.2% groups were well balanced and there were no significant baseline after 4 h, followed by a slowly declining plateau, differences among the three groups in demographic data the increase in sGaw after 12 h being 255.6% baseline. and baseline lung function. Medication usage was comp- The combination of salmeterol and ipratropium led to a arable among the three groups. In total, 20 patients were maximum increase in sGaw of 948.9% baseline after 2 h, withdrawn from the study during treatment, eight in the the increase after 12 h also being 255.1% baseline. placebo group, seven in the salmeterol group and five in The mean AUCs of the parameters investigated (FEV1, the salmeterol plus ipratropium group (NS). The reasons FVC and sGaw) were significantly greater in both the for withdrawal were: exacerbation of COPD (three pa- salmeterol and the salmeterol plus ipratropium groups than tients in the salmeterol group, two in the salmeterol plus in the placebo group (p<0.001). Furthermore, the AUCs of ipratropium group and four in the placebo group), adverse FEV1 and sGaw after salmeterol plus ipratropium, com- events not considered to be drug related (four patients), pared with those after salmeterol, were in favour of the missed visits (one patient), poor compliance (one patient), former group (p<0.05). The AUCs of FVC did not sig- and others (five patients). The data of the withdrawn pa- nificantly differ between the salmeterol and salmeterol plus tients were included in the (intention-to-treat) analysis. ipratropium groups. Considering each single time point after inhalation, there were no longer any significant dif- First-dose study ferences in FEV1, FVC and sGaw between the salmeterol The time/response curves after the first dose of salme- and the salmeterol plus ipratropium groups after 6 h. There terol and salmeterol plus ipratropium over a period of 12 h were no significant treatment effects on cardiac frequency after inhalation are shown in figure 1. There were no sig- and blood pressure. nificant differences in baseline FEV1, FVC and sGaw among the three groups. After inhalation of salmeterol there was a meanSEM peak increase in FEV1 of 70.7% Twelve-week treatment pred after 2 h, followed by a plateau. After 12 h, the im- Day- and night-time symptom scores. The mean daytime provement was still 21% pred. Salmeterol plus ipra- symptom score during the run-in period did not differ tropium produced a peak increase in FEV1 of 110.8% among the three groups (fig. 2). Throughout the treatment pred after 2 h. After 12 h, the improvement was 30.8% period there was a meanSEM decrease from 1.90.1 to pred. The improvement in FVC in the two active treat- 1.70.1 in the placebo group NS, from 2.00.1 to 1.40.1 ment groups was in line with that in FEV1. After salme- (p<0.001) in the salmeterol group and from 2.00.1 to SALMETEROL VERSUS SALMETEROL PLUS IPRATROPIUM IN COPD 881 a) treatment, the percentages increased to 493.0 for the sal- 12.5 meterol group, 573.0 for the salmeterol plus ipratropium group and 344.0 for the placebo group, all changes being 10.0 significant (p<0.05). Improvements in both active treatment arms were better than in the placebo group (p<0.05), 7.5 ■ ■ ■ ■ ■ but no significant difference was observed between the 5.0 ■ ■ salmeterol and the salmeterol plus ipratropium groups. The % pred ■ 1 meanSEM night-time symptom score during the run-in ■ ● 2.5 ● ● ■ period was 0.70.1 in the placebo group, 0.80.1 in the FEV ● ● ∆ ● salmeterol group and 0.80.1 in the salmeterol plus ipra- ●■ ● 0.0 ● tropium group. No change was found in the mean scores ● -2.5 ● during treatment for all groups. The percentage of nights without symptoms at baseline was not different among the -5.0 three groups, 526.0 in the placebo group, 545.9 in the salmeterol group and 506.4 in the salmeterol plus ipra- b) tropium group. No significant changes were observed in 20 these percentages during treatment for all groups. 16 Rescue medication. The patients were permitted to use sal- ■ 12 ■ ■ ■ butamol as rescue medication during both the run-in and ■ ■ ■ the treatment periods. Compared with placebo, treatment 8 ■ with both salmeterol and salmeterol plus ipratropium were ● ● ■ ● ■ associated with a higher percentage of days and nights 4 ● ● FVC % pred ● without use of additional salbutamol (p<0.01) (fig. 2). No ∆ ● ●■ ● difference was observed between the two active treatment 0 ● ● arms (p=0.35). During the run-in period, the meanSEM -4 percentages of days with additional use of salbutamol -8 were 981.8, 972.2 and 933.2 for placebo, salmeterol and salmeterol plus ipratropium, respectively. During the c) treatment period, these values decreased to 745.0, 34 0.7 5.5 and 275.5, respectively. At baseline, the percentages of nights with rescue bronchodilator usage were 37 6.1, 0.6 376.3 and 506.9 for placebo, salmeterol and salmeterol plus ipratropium, respectively. During the treatment per- -1

·s ■ ■ iod these percentages were reduced to 336.3, 172.9 and ■ -1 0.5 ■ ■ 244.1%, respectively. ■ ■

kPa ■ ■

aw ■ Peak expiratory flow measurements. Improvements in 0.4 sG morning PEFs were significantly better in both active ■ ● ● ● ● treatment groups than in the placebo group (p<0.001), ● ● whereas no difference was observed between the salmeterol 0.3 ● ● ● ● ● and the salmeterol plus ipratropium groups (fig. 3). During the treatment period, meansem morning PEF changed 0.2 from 2389 L.min-1 to 2369 L.min-1 in the placebo 0123456789101112 group, from 2469 to 26211 L.min-1 in the salmeterol -1 Time h group and from 25211 to 27712 L.min in the salmeterol plus ipratropium group. Also, the changes in evening Fig. 1. ± Change (D) in: a) forced expiratory volume in one second (FEV1); b) forced vital capacity (FVC); and c) absolute values of PEFs (measured before the evening dose of salmeterol specific airway conductance (sGaw) during 12 h after salmeterol (&), but after the 17:00 h dose of ipratropium) were in favour salmeterol plus ipratropium (u) and placebo (*). Data are presented as of both active treatment arms compared with placebo meanSEM. % pred: percentage of the predicted value. (p<0.001), whereas the improvement was better in the salmeterol plus ipratropium group than in the salmeterol 1.30.1 (p<0.001) in the salmeterol plus ipratropium group (p<0.01). Evening PEF changed from 25910 to group. There was a significant difference in the change 25310 L.min-1 in the placebo group, from 25710 to from the run-in period in the daytime symptom score 27111 L.min-1 in the salmeterol group and from 27110 between the salmeterol and the placebo groups (p<0.005) to 29711 L.min-1 in the salmeterol plus ipratropium and between the salmeterol plus ipratropium and placebo group. groups (p<0.001), but not between the salmeterol and the salmeterol plus ipratropium groups. Clinic lung function. During the 12-week treatment per- In addition, analysis was made of the meanSEM percen- iod, the meanSEM increase in FEV1 was 10.9% pred for tage of days with a minimal level of symptoms (score #1). the placebo, 50.9% pred for the salmeterol and 80.8% During the run-in period, this amounted 142.9% in the pred for the salmeterol plus ipratropium treatment groups salmeterol group, 143.0% in the salmeterol plus ipratro- (fig. 4). All differences between the three groups were pium group and 173.7% in the placebo group. During significant (p<0.01). The change in FVC was 41.2% 882 J.A. VAN NOORD ET AL.

a) b) 80 70 70 60 ■ ■ ● ■ 60 50 ●■ 50 ■ ● ● ■ ■ 40 40 ● ● 30 30 ● 20 20 ● ■

10 Nights without symptoms % 10 Days with minimal symptoms % Days 0 0 c) d) 120 50

100 ●■ 40 ●■ ● ● 80 ● ● 30 ● ● 60 20 40 ■ ■ ■ 10 ■ 20 ■ ■ Days with rescue medication % Days 0 Nights with rescue medication % 0 0 1–4 4–8 8–12 0 1–4 4–8 8–12 Weeks Weeks Fig. 2. ± Percentage of: a) days with minimal (score #1) symptoms; b) nights without symptoms; c) days with rescue medication; and d) nights with rescue medication per 4-week period before and during treatment with salmeterol (&), salmeterol plus ipratropium (u) and placebo (*). Data are presented as meanSEM. pred after placebo, 71.2% pred after salmeterol and possible and probably drug-related side-effects were simi- 121.2% after salmeterol plus ipratropium. The differen- lar among the three groups. The most common drug-related ces between salmeterol plus ipratropium versus salmeterol adverse events were headache (six patients in the salmeterol alone and between salmeterol plus ipratropium versus pla- plus ipratropium group, four in the salmeterol group and 11 cebo were both significant (p<0.01), whereas there was no in the placebo group) and cough after inhalation (two significant difference between the change in FVC after patients in the salmeterol plus ipratropium group, three in placebo and salmeterol (p=0.055). The meanSEM incr- the salmeterol group and three in the placebo group). In the ease in sGaw from baseline was 166% after placebo, salmeterol plus ipratropium group, one patient reported 366% after salmeterol and 616% after salmeterol plus muscle cramps. During the 12-week treatment period, 35 ipratropium, all differences between the three groups patients experienced a COPD exacerbation, 18 (36%) in the being significant (p<0.01). The maximum improvement placebo group, 11 (23%) in the salmeterol group and six in FEV1, FVC and sGaw was reached at visit 3 after 4 (13%) in the salmeterol plus ipratropium group. The only weeks of treatment and was sustained throughout the significant difference was that between the salmeterol plus treatment period in the salmeterol and salmeterol plus ipratropium group and the placebo group (p<0.01). ipratropium groups. In contrast, in the placebo group, FEV1, FVC and sGaw continued to improve, albeit at a lower level, during the treatment period. At visits 3, 4 and 5, Discussion i.e. after 4, 8 and 12 weeks of treatment, inhalation of salbutamol 200 mg produced an additional improvement in At the start of the 12-week treatment period, the time- FEV1 in all groups (fig. 5). This increase in FEV1 post- response curve was evaluated after the first dose of the salbutamol was greater in the placebo than in the salmet- active treatment and placebo regimens. The results demon- erol and salmeterol plus ipratropium groups (p<0.01). strate that, in patients with stable COPD and severe airflow There were no significant differences in mean FEV1 after obstruction, 50 mg of salmeterol produces a significant salbutamol at any visit among the three groups. In other improvement in lung function as assessed by FEV1 and words, active treatment with salmeterol or salmeterol plus sGaw, with a peak effect after 2±4 h and a duration of action ipratropium did not change the maximum attainable of $12 h. In addition, the combination of salmeterol 50 mg FEV1 after salbutamol. and ipratropium bromide 40 mg elicited a greater bronchodilator response than salmeterol alone during the first 6 h Adverse events and exacerbations. Both active treatments after inhalation. The present results with salmeterol are in were well tolerated. The reported incidence and nature of line with those of CAZZOLA et al. [4]. These authors, SALMETEROL VERSUS SALMETEROL PLUS IPRATROPIUM IN COPD 883 a) lung function after the morning dose and by recording PEF 310 at home before the morning dose. There was a significant improvement in clinic lung function as measured by FEV1 and sGaw in both the salmeterol and the salmeterol plus ip- -1 290 ratropium groups. This improvement was greater in the salmeterol plus ipratropium group than in the salmeterol 270 group, indicating that, in these patients with severe airflow ■ ■ ■ obstruction, the recommended dose of salmeterol alone 250 was not able to achieve maximum bronchodilation and that ■ combination of the two drugs had an additive effect. More- ● ● ● Morning PEF L·min ● over, it turned out that, even after salmeterol plus ipra- 230 tropium at the conventional dose further bronchodilation was still possible with salbutamol. These results are in 210 keeping with those of CAZZOLA et al. [19] who showed that pretreatment with a conventional dose of salmeterol in b) COPD does not preclude the possibility of inducing fur- 330 ther bronchodilation with salbutamol. The hypothesis that higher doses of salmeterol or ipratropium alone could have 310 induced an improvement in airflow obstruction similar to -1 that following the combination of salmeterol and ipratro- 290 pium was not tested. There are several studies supporting

■ this hypothesis. EASTON et al. [20] demonstrated in COPD 270 ■ ■ that, when either salbutamol or ipratropium was given ● ■ ● ● alone at the maximal dose, both drugs were equally effect- 250 ● ive and the addition of the second drug did not cause further

Evening PEF L·min Evening bronchodilation. IKEDA et al. [21] found that, in patients 230 with COPD, 160 or 240 mg ipratropium produced a greater improvement in FEV1 than 40 or 80 mg. In line with these 210 results, it was found that there was no difference in the 0 1–4 4–8 8–12 maximum achievable FEV1 after salbutamol in the two Weeks treatment arms. It generally follows that the risk of side- Fig. 3. ± Peak expiratory flow (PEF) in: a) the morning; and b) the effects increases with increasing doses of any drug. The evening over 4-week periods before and during treatment with most important rationale for combination therapy is there- salmeterol (&), salmeterol plus ipratropium (u) and placebo (*). fore probable very favourable ratio of efficacy and side- Data are presented as meanSEM. effects. The incidence of side-effects in the present study during combination treatment was very low indeed. comparing single doses of 25, 50 and 75 mg salmeterol in The present study also demonstrated that, in both active patients with COPD, found that all doses induced a treatment groups, improvement in clinic lung function was significant improvement in FEV1 over the 12 h maximal after 4 weeks of treatment and remained stable monitoring period. As the results after 50 and 75 mgof throughout the complete course of the treatment period, salmeterol were similar, they suggested that 50 mg was confirming that long-term use of salmeterol with or without the best dose of salmeterol in these patients. It has been ipratropium does not lead to tolerance of clinical importance shown that this recommended dose of 50 mg salmeterol to this long-acting b2-agonist. Moreover, during treatment produces a similar [11] or even greater improvement [16] there was no change in the bronchodilator response to in airway obstruction than the recommended dose of 40 salbutamol. mg ipratropium bromide. To date, only one clinical trial Furthermore, only the combination treatment of salme- has been carried out on the efficacy of combined terol plus ipratropium produced a significant increase in administration of salmeterol and ipratropium in COPD. FVC, which can be interpreted as an improvement in air MATERA et al. [17] found that, in an ill-defined group of trapping and a diminution of airway closure. The result 12 patients with COPD, combination of salmeterol and was not expected apriorias it is generally thought that ipratropium in the clinically recommended doses did not ipratropium produces its bronchodilating effect mainly in produce greater bronchodilation than salmeterol alone. the central airways [22, 23]. However, a number of large clinical trials indicate that Unlike clinic lung function, which was measured after combination of the short-acting b2-agonist albuterol and the morning dose of the study drugs, morning PEF was ipratropium produces better improvement in spirometric measured at home before the morning dose of the study indices than either drug taken alone at the recommended medication. Both salmeterol and salmeterol plus ipratro- doses [12, 18]. In contrast to the findings of MATERA et al. pium produced increases in morning PEF, which were [17], the present study demonstrated that this was also greater than those after placebo but did not differ from each true for the combination of the long-acting b2-agonist other. Considering the duration of action of ipratropium of salmeterol and ipratropium during the first 6 h after ~6±8 h, the long duration of action of the evening dose of dosing, after which time the additive effect of the short- salmeterol must account for this increase in trough values acting ipratropium presumably diminished. of PEF in the morning. This improvement in morning Changes in lung function during the 12-week treatment PEF remained stable throughout the 12-week treatment period were evaluated in two ways, by measuring clinic period, indicating an absence of tolerance to the duration 884 J.A. VAN NOORD ET AL.

tances and symptoms are unrelated to change in FEV1 aft- a) ** 10 ** er bronchodilator use and routine reversibility testing is ** not a good predictor of symptomatic benefit in COPD. 8 Several authors have shown that small or even no changes in FEV1 after bronchodilation may nevertheless be as- 6 sociated with an improvement in symptoms and clinic- ■ ■ ally significant gains in health and well-being [3, 6, 8].

% pred ■ 1 4 WEGNER et al. [24] found that severe COPD could be characterized by three statistically independent entities: 1) FEV

∆ 2 ● ● exercise capacity, dyspnoea and quality-of-life scores; 2) airflow obstruction; and 3) pulmonary hyperinflation. It is ● 0 ●■ obvious that improvement in airways obstruction is not the most relevant end point of this study and that im- -2 proving health and well-being is far more important for b) the patient. Therefore, the primary outcome parameters of 14 ** the present study were daytime and night-time symptom ** scores, use of rescue medication and exacerbation rate. It 12 turned out that treatment with both salmeterol and salme- ** terol plus ipratropium resulted in significant reductions 10 in daytime symptoms, an increase in days with minimal symptoms and a decrease in the use of rescue sal- 8 ■ butamol during the day and the night. In contrast to the ■ ■ ● results of the lung function measurements at home and in

FVC % pred 6

∆ the clinic, there were no significant differences between ● 4 the two active treatment groups. Night-time symptom ● scores were also recorded, but no difference between the 2 placebo and active treatment groups could be detected. The reason may be that night-time symptom scores were 0 ■● already very low during the run-in period, leaving little room for improvement during the treatment period. The c) data on exacerbation rates in this study must be inter- 0.6 ** preted with caution as the treatment was relatively short. ** However, it turned out that the number of exacerbations ** was highest in the placebo group and lowest in the sal-

-1 0.5

·s meterol plus ipratropium group. The patients in this study

-1 ■ also completed the Chronic Respiratory Questionnaire ■ ■ kPa and the St George's Respiratory Questionnaire (SGRQ), 0.4 aw ● the results of which were published very recently [25]. It G ■ ● s suffices to note here that the only domain that showed an ● improvement of clinical relevance was the SGRQ sym- 0.3 ● ptoms domain in patients receiving salmeterol plus ipratropium bromide. From the literature, little is known 0.2 ■● about the relationship between diary card scores and health status questionnaires in COPD. Diary cards are 04812suited to close monitoring of the disease and daily appli- Weeks cation, whereas questionnaires provide intermittent health Fig. 4. ± Change (D) in: a) forced expiratory volume in one second estimates [26]. In contrast to questionnaires, diary cards (FEV1); b) forced vital capacity (FVC); and c) absolute values of have not been validated and tested intensively [26]. This specific airway conductance (sGaw) after 4, 8 and 12 weeks of treatment study demonstrates that diary cards including symptom with salmeterol (&), salmeterol plus ipratropium (u), and placebo (*). scores and use of rescue medication correlate imperfectly Data are presented as meanSEM. % pred: percentage of the predicted value. **: p<0.01 versus salmeterol alone. with health status questionnaires. In conclusion, in patients with chronic obstructive pulmonary disease and severe airflow obstruction long-term of the bronchodilating effect of salmeterol. Analysis of the treatment both with salmeterol and salmeterol plus ipra- evening PEF has little meaning, because measurements tropium is effective and safe. There was added benefit of were performed before the evening dose of salmeterol but combination therapy with salmeterol plus ipratropium com- only 2 h after the inhalation of the evening dose of pared with treatment with salmeterol alone in terms of ipratropium. improvement in airways obstruction, but not in terms of It is known that in COPD relations between lung func- improvement in symptom control or in use of rescue sal- tion measurements, symptoms, exercise capacity, walking butamol. In addition, there was some evidence that active distance and quality of life are not straightforward. HAY et treatment, especially with combination therapy, decreased al. [3] demonstrated that improvements in walking dis- the rate of exacerbations. SALMETEROL VERSUS SALMETEROL PLUS IPRATROPIUM IN COPD 885

8. Grove A, Lipworth BJ, Reid P, et al. Effects of regular a) salmeterol on lung function and exercise capacity in patients 1.7 with obstructive airways disease. Thorax 1996; 51: 689±693. 9. Boyd G, Morice AH, Pounsford JC, et al. An evaluation of salmeterol in the treatment of chronic obstructive pulmonary 1.6 disease (COPD). Eur Respir J 1997; 10: 815±821. 10. Jones PW, Bosh TK. Quality of life changes in COPD

L patients treated with salmeterol. Am J Respir Crit Care Med 1 1.5 1997; 155: 1283±1289.

FEV ■ 11. Mahler DA, Donohue JF, Barbee RA, et al. Efficacy of ● ■ ■ ● ● salmeterol xinafoate in the treatment of COPD. Chest 1999; ● ■ 115: 957±965. 1.4 12. Combivent Inhalation Aerosol Study Group. In chronic obstructive pulmonary disease, a combination of ipratropium and albuterol is more effective than either agent alone. 1.3 Chest 1994; 105: 1411±1419. 13. Palmer JBD, Stuart AM, Shepherd GL, et al. Inhaled b) salmeterol in the treatment of patients with moderate-to- 10 severe reversible obstructive airways disease: a three month comparison of the efficacy and safety of twice daily salmeterol (100 mg) with salmeterol (50 mg). Respir Med 1992; 8 86: 409±417. ● ● ● 14. Schluchter MD. Module 5V. In: Dixon WJ, ed. BMDP Statistical Software Manual. Berkeley, University of Cali- 6 ●■ fornia Press, 1990: 1207±1244. % pred 1 15. Altman DG. Practical Statistics for medical research. Lon- ■ don, Chapman & Hall, 1991: p. 211. FEV 4 ■

∆ 16. Patakas D, Andreadis D, Mavrofridis E, et al. Comparison of the effects of salmeterol and ipratropium bromide on ■ 2 exercise performance and breathlessness in patients with stable chronic obstructive pulmonary disease. Respir Med 1998; 92: 1116±1121. 0 17. Matera MG, Caputi M, Cazzola M. A combination with -24 8 12 clinical recommended dosages of salmeterol and ipratropium is not more effective than salmeterol alone in patients Time on treatment weeks with chronic obstructive pulmonary disease. Respir Med Fig. 5. ± a) Forced expiratory volume in one second (FEV1); and b) 1996; 90: 497±499. change (D) in FEV1 after salbutamol at the start of the run-in period and 18. Wilson JD, Serby CW, Menjoge SS, et al. The efficacy and during treatment with salmeterol (&), salmeterol plus ipratropium (u), safety of combination bronchodilator therapy. Eur Respir Rev and placebo (*) after 4, 8 and 12 weeks of treatment. Data are presented 1996; 6: 39, 286±289. as meanSEM. % pred: percentage of the predicted value. 19. Cazzola M, Di Perna F, Noschese P, et al. Effects of formoterol, salmeterol or oxitropium bromide on airway responses to salbutamol in COPD. Eur Respir J 1998; 11: 1337±1341. References 20. Easton PA, Jadue C, Dhingra S, et al. A comparison of the bronchodilating effect of a b -adrenergic agent (albuterol) 1. American Thoracic Society. Standards for the diagnosis and 2 and an anticholinergic agent (ipratropium bromide), given by care of patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1995; 152: S77±S120. aerosol alone or in sequence. N Engl J Med 1986; 315: 735± 2. Siafakas NM, Vermeire P, Pride NB, et al. Optimal as- 739. sessment and management of chronic obstructive pulmonary 21. Ikeda A, Nishimura K, Koyama H, et al. Dose response study disease (COPD): ERS consensus statement. Eur Respir J of ipratropium bromide aerosol on maximum exercise per- 1995; 8: 1398±1420. formance in stable patients with chronic obstructive pul- 3. Hay JG, Stone P, Carter J, et al. Bronchodilator reversibility, monary disease. Thorax 1996; 51: 48±53. exercise performance and breathlessness in stable chronic 22. Ingram RH, Wellman JJ, Mc Fadden ER, et al. Relative obstructive pulmonary disease. Eur Respir J 1992; 5: 659± contribution of large and small airways to flow limitation in 664. normal subjects before and after atropine and isoproterenol. J 4. Cazzola M, Matera MG, Santangelo G, et al. Salmeterol and Clin Invest 1977; 59: 696±703. formoterol in partially reversible severe chronic obstructive 23. Barnes PJ, Basbaum CB, Nadel JA. Autoradiographic loc- pulmonary disease: a dose-response study. Respir Med 1995; alization of autonomic receptors in airway smooth muscle. 89: 357±362. Marked differences between large and small airways. Am Rev 5. Cazzola M, Santangelo G, Piccolo A, et al. Effect of Respir Dis 1983; 127: 758±762. salmeterol and formoterol in patients with chronic obstr- 24. Wegner RE, Jorres RA, Kirsten DK, et al. Factor analysis of uctive pulmonary disease. Pulm Pharmacol 1994; 7: 103± exercise capacity, dyspnoea ratings and lung function in 107. patients with severe COPD. Eur Respir J 1994; 7: 725±729. 6. Cazzola M, Matera MG, Di Pema F, et al. A comparison of 25. Rutten-van MoÈlken M, Roos B, Van Noord JA. An em- bronchodilating effects of salmeterol and oxitropium bro- pirical comparison of the St George's Respiratory Ques- mide in stable chronic obstructive pulmonary disease. Respir tionnaire (SGRQ) and the Chronic Respiratory Disease Med 1998; 92: 354±357. Questionnaire (CRQ) in a clinical trial setting. Thorax 1999; 7. Ulrik CS. Efficacy of inhaled salmeterol in the management 54: 995±1003. of smokers with chronic obstructive pulmonary disease: a 26. Barley EA, Jones PW. A comparison of global questions single centre randomised, double blind, placebo controlled versus health status questionnaires as measures of the sev- crossover study. Thorax 1995; 50: 750±754. erity and impact of asthma. Eur Respir J 1999; 14: 591±596.